Scholarlink

Modern LLM models such as BERT, ChatGPT, DeepSeek have shown great potential in solving various tasks, including text classification, text generation, analysis and summary of documents. In this paper, we show that these models close to classical ML approaches based on decision trees not only in text processing, but also in processing classical tabular data (quantitative and categorical) using the example of solving the faculty forecasting problem for university applicants. Based on the dataset we prepared from 1597 applicants, the LLM models showed fairly close results to classical machine learning methods.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Rewriting the Rules: LLMs Vs. Traditional ML in University Admissions

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ivan Chepikov,
  • Ilia Karpov

摘要

Modern LLM models such as BERT, ChatGPT, DeepSeek have shown great potential in solving various tasks, including text classification, text generation, analysis and summary of documents. In this paper, we show that these models close to classical ML approaches based on decision trees not only in text processing, but also in processing classical tabular data (quantitative and categorical) using the example of solving the faculty forecasting problem for university applicants. Based on the dataset we prepared from 1597 applicants, the LLM models showed fairly close results to classical machine learning methods.